Training in progress, step 1728, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 891644712
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8d157e7286a6dfc4ba5aacde53caecbbb37a38b4599aefda7a1cea8fa2b162ea
|
| 3 |
size 891644712
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1783444794
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:96ad3332c7a20eba91ce157eec436b4cce08960829f029b3ab7446b861d4f3da
|
| 3 |
size 1783444794
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:edb897ff7c5edb0daaf53db6c8527f7da45dd70041d3faeafb18eb1d69b53ca7
|
| 3 |
size 14244
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c92509254bae5d36eb0a64d8994d4ff09c04cd0249ac08027b0f40c16d4a82bf
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
@@ -5607,6 +5607,454 @@
|
|
| 5607 |
"learning_rate": 0.000122927204684282,
|
| 5608 |
"loss": 0.3386,
|
| 5609 |
"step": 1600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5610 |
}
|
| 5611 |
],
|
| 5612 |
"logging_steps": 2,
|
|
@@ -5626,7 +6074,7 @@
|
|
| 5626 |
"attributes": {}
|
| 5627 |
}
|
| 5628 |
},
|
| 5629 |
-
"total_flos":
|
| 5630 |
"train_batch_size": 8,
|
| 5631 |
"trial_name": null,
|
| 5632 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9584026622296173,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1728,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 5607 |
"learning_rate": 0.000122927204684282,
|
| 5608 |
"loss": 0.3386,
|
| 5609 |
"step": 1600
|
| 5610 |
+
},
|
| 5611 |
+
{
|
| 5612 |
+
"epoch": 0.8885191347753744,
|
| 5613 |
+
"grad_norm": 0.3254069983959198,
|
| 5614 |
+
"learning_rate": 0.00012275228026094881,
|
| 5615 |
+
"loss": 0.5074,
|
| 5616 |
+
"step": 1602
|
| 5617 |
+
},
|
| 5618 |
+
{
|
| 5619 |
+
"epoch": 0.8896283971159179,
|
| 5620 |
+
"grad_norm": 0.37417536973953247,
|
| 5621 |
+
"learning_rate": 0.00012257728238736467,
|
| 5622 |
+
"loss": 0.5318,
|
| 5623 |
+
"step": 1604
|
| 5624 |
+
},
|
| 5625 |
+
{
|
| 5626 |
+
"epoch": 0.8907376594564614,
|
| 5627 |
+
"grad_norm": 0.35727638006210327,
|
| 5628 |
+
"learning_rate": 0.000122402211628468,
|
| 5629 |
+
"loss": 0.4729,
|
| 5630 |
+
"step": 1606
|
| 5631 |
+
},
|
| 5632 |
+
{
|
| 5633 |
+
"epoch": 0.891846921797005,
|
| 5634 |
+
"grad_norm": 0.26733312010765076,
|
| 5635 |
+
"learning_rate": 0.00012222706854943255,
|
| 5636 |
+
"loss": 0.421,
|
| 5637 |
+
"step": 1608
|
| 5638 |
+
},
|
| 5639 |
+
{
|
| 5640 |
+
"epoch": 0.8929561841375485,
|
| 5641 |
+
"grad_norm": 0.22087961435317993,
|
| 5642 |
+
"learning_rate": 0.00012205185371566554,
|
| 5643 |
+
"loss": 0.3354,
|
| 5644 |
+
"step": 1610
|
| 5645 |
+
},
|
| 5646 |
+
{
|
| 5647 |
+
"epoch": 0.894065446478092,
|
| 5648 |
+
"grad_norm": 0.4256139099597931,
|
| 5649 |
+
"learning_rate": 0.00012187656769280578,
|
| 5650 |
+
"loss": 0.4,
|
| 5651 |
+
"step": 1612
|
| 5652 |
+
},
|
| 5653 |
+
{
|
| 5654 |
+
"epoch": 0.8951747088186356,
|
| 5655 |
+
"grad_norm": 0.2818162441253662,
|
| 5656 |
+
"learning_rate": 0.00012170121104672196,
|
| 5657 |
+
"loss": 0.4098,
|
| 5658 |
+
"step": 1614
|
| 5659 |
+
},
|
| 5660 |
+
{
|
| 5661 |
+
"epoch": 0.8962839711591791,
|
| 5662 |
+
"grad_norm": 0.2936331331729889,
|
| 5663 |
+
"learning_rate": 0.00012152578434351071,
|
| 5664 |
+
"loss": 0.436,
|
| 5665 |
+
"step": 1616
|
| 5666 |
+
},
|
| 5667 |
+
{
|
| 5668 |
+
"epoch": 0.8973932334997227,
|
| 5669 |
+
"grad_norm": 0.2814910113811493,
|
| 5670 |
+
"learning_rate": 0.00012135028814949487,
|
| 5671 |
+
"loss": 0.4096,
|
| 5672 |
+
"step": 1618
|
| 5673 |
+
},
|
| 5674 |
+
{
|
| 5675 |
+
"epoch": 0.8985024958402662,
|
| 5676 |
+
"grad_norm": 0.3062569797039032,
|
| 5677 |
+
"learning_rate": 0.00012117472303122157,
|
| 5678 |
+
"loss": 0.4595,
|
| 5679 |
+
"step": 1620
|
| 5680 |
+
},
|
| 5681 |
+
{
|
| 5682 |
+
"epoch": 0.8996117581808097,
|
| 5683 |
+
"grad_norm": 0.3199828565120697,
|
| 5684 |
+
"learning_rate": 0.00012099908955546044,
|
| 5685 |
+
"loss": 0.4696,
|
| 5686 |
+
"step": 1622
|
| 5687 |
+
},
|
| 5688 |
+
{
|
| 5689 |
+
"epoch": 0.9007210205213533,
|
| 5690 |
+
"grad_norm": 0.35935017466545105,
|
| 5691 |
+
"learning_rate": 0.00012082338828920185,
|
| 5692 |
+
"loss": 0.5822,
|
| 5693 |
+
"step": 1624
|
| 5694 |
+
},
|
| 5695 |
+
{
|
| 5696 |
+
"epoch": 0.9018302828618968,
|
| 5697 |
+
"grad_norm": 0.2030808925628662,
|
| 5698 |
+
"learning_rate": 0.00012064761979965497,
|
| 5699 |
+
"loss": 0.3524,
|
| 5700 |
+
"step": 1626
|
| 5701 |
+
},
|
| 5702 |
+
{
|
| 5703 |
+
"epoch": 0.9029395452024404,
|
| 5704 |
+
"grad_norm": 0.29535773396492004,
|
| 5705 |
+
"learning_rate": 0.00012047178465424596,
|
| 5706 |
+
"loss": 0.3698,
|
| 5707 |
+
"step": 1628
|
| 5708 |
+
},
|
| 5709 |
+
{
|
| 5710 |
+
"epoch": 0.9040488075429839,
|
| 5711 |
+
"grad_norm": 0.26572179794311523,
|
| 5712 |
+
"learning_rate": 0.00012029588342061621,
|
| 5713 |
+
"loss": 0.3789,
|
| 5714 |
+
"step": 1630
|
| 5715 |
+
},
|
| 5716 |
+
{
|
| 5717 |
+
"epoch": 0.9051580698835274,
|
| 5718 |
+
"grad_norm": 0.4271789491176605,
|
| 5719 |
+
"learning_rate": 0.00012011991666662044,
|
| 5720 |
+
"loss": 0.5669,
|
| 5721 |
+
"step": 1632
|
| 5722 |
+
},
|
| 5723 |
+
{
|
| 5724 |
+
"epoch": 0.906267332224071,
|
| 5725 |
+
"grad_norm": 0.35716575384140015,
|
| 5726 |
+
"learning_rate": 0.00011994388496032487,
|
| 5727 |
+
"loss": 0.4521,
|
| 5728 |
+
"step": 1634
|
| 5729 |
+
},
|
| 5730 |
+
{
|
| 5731 |
+
"epoch": 0.9073765945646145,
|
| 5732 |
+
"grad_norm": 0.2956486642360687,
|
| 5733 |
+
"learning_rate": 0.00011976778887000543,
|
| 5734 |
+
"loss": 0.3755,
|
| 5735 |
+
"step": 1636
|
| 5736 |
+
},
|
| 5737 |
+
{
|
| 5738 |
+
"epoch": 0.908485856905158,
|
| 5739 |
+
"grad_norm": 0.3578818738460541,
|
| 5740 |
+
"learning_rate": 0.0001195916289641459,
|
| 5741 |
+
"loss": 0.4935,
|
| 5742 |
+
"step": 1638
|
| 5743 |
+
},
|
| 5744 |
+
{
|
| 5745 |
+
"epoch": 0.9095951192457016,
|
| 5746 |
+
"grad_norm": 0.3232196867465973,
|
| 5747 |
+
"learning_rate": 0.00011941540581143608,
|
| 5748 |
+
"loss": 0.4826,
|
| 5749 |
+
"step": 1640
|
| 5750 |
+
},
|
| 5751 |
+
{
|
| 5752 |
+
"epoch": 0.9107043815862451,
|
| 5753 |
+
"grad_norm": 0.2944696247577667,
|
| 5754 |
+
"learning_rate": 0.00011923911998076988,
|
| 5755 |
+
"loss": 0.3827,
|
| 5756 |
+
"step": 1642
|
| 5757 |
+
},
|
| 5758 |
+
{
|
| 5759 |
+
"epoch": 0.9118136439267887,
|
| 5760 |
+
"grad_norm": 0.27748194336891174,
|
| 5761 |
+
"learning_rate": 0.00011906277204124363,
|
| 5762 |
+
"loss": 0.5143,
|
| 5763 |
+
"step": 1644
|
| 5764 |
+
},
|
| 5765 |
+
{
|
| 5766 |
+
"epoch": 0.9129229062673322,
|
| 5767 |
+
"grad_norm": 0.2819176912307739,
|
| 5768 |
+
"learning_rate": 0.00011888636256215413,
|
| 5769 |
+
"loss": 0.4159,
|
| 5770 |
+
"step": 1646
|
| 5771 |
+
},
|
| 5772 |
+
{
|
| 5773 |
+
"epoch": 0.9140321686078757,
|
| 5774 |
+
"grad_norm": 0.3712371289730072,
|
| 5775 |
+
"learning_rate": 0.00011870989211299686,
|
| 5776 |
+
"loss": 0.5419,
|
| 5777 |
+
"step": 1648
|
| 5778 |
+
},
|
| 5779 |
+
{
|
| 5780 |
+
"epoch": 0.9151414309484193,
|
| 5781 |
+
"grad_norm": 0.2566871643066406,
|
| 5782 |
+
"learning_rate": 0.00011853336126346406,
|
| 5783 |
+
"loss": 0.4926,
|
| 5784 |
+
"step": 1650
|
| 5785 |
+
},
|
| 5786 |
+
{
|
| 5787 |
+
"epoch": 0.9162506932889628,
|
| 5788 |
+
"grad_norm": 0.38075196743011475,
|
| 5789 |
+
"learning_rate": 0.0001183567705834431,
|
| 5790 |
+
"loss": 0.4784,
|
| 5791 |
+
"step": 1652
|
| 5792 |
+
},
|
| 5793 |
+
{
|
| 5794 |
+
"epoch": 0.9173599556295063,
|
| 5795 |
+
"grad_norm": 0.30149558186531067,
|
| 5796 |
+
"learning_rate": 0.00011818012064301433,
|
| 5797 |
+
"loss": 0.3791,
|
| 5798 |
+
"step": 1654
|
| 5799 |
+
},
|
| 5800 |
+
{
|
| 5801 |
+
"epoch": 0.9184692179700499,
|
| 5802 |
+
"grad_norm": 0.37024736404418945,
|
| 5803 |
+
"learning_rate": 0.00011800341201244954,
|
| 5804 |
+
"loss": 0.4495,
|
| 5805 |
+
"step": 1656
|
| 5806 |
+
},
|
| 5807 |
+
{
|
| 5808 |
+
"epoch": 0.9195784803105934,
|
| 5809 |
+
"grad_norm": 0.30697816610336304,
|
| 5810 |
+
"learning_rate": 0.00011782664526220992,
|
| 5811 |
+
"loss": 0.385,
|
| 5812 |
+
"step": 1658
|
| 5813 |
+
},
|
| 5814 |
+
{
|
| 5815 |
+
"epoch": 0.920687742651137,
|
| 5816 |
+
"grad_norm": 0.48160433769226074,
|
| 5817 |
+
"learning_rate": 0.00011764982096294432,
|
| 5818 |
+
"loss": 0.3435,
|
| 5819 |
+
"step": 1660
|
| 5820 |
+
},
|
| 5821 |
+
{
|
| 5822 |
+
"epoch": 0.9217970049916805,
|
| 5823 |
+
"grad_norm": 0.3319704234600067,
|
| 5824 |
+
"learning_rate": 0.00011747293968548734,
|
| 5825 |
+
"loss": 0.4893,
|
| 5826 |
+
"step": 1662
|
| 5827 |
+
},
|
| 5828 |
+
{
|
| 5829 |
+
"epoch": 0.922906267332224,
|
| 5830 |
+
"grad_norm": 0.2384938895702362,
|
| 5831 |
+
"learning_rate": 0.00011729600200085752,
|
| 5832 |
+
"loss": 0.4826,
|
| 5833 |
+
"step": 1664
|
| 5834 |
+
},
|
| 5835 |
+
{
|
| 5836 |
+
"epoch": 0.9240155296727676,
|
| 5837 |
+
"grad_norm": 0.2830295264720917,
|
| 5838 |
+
"learning_rate": 0.00011711900848025555,
|
| 5839 |
+
"loss": 0.5185,
|
| 5840 |
+
"step": 1666
|
| 5841 |
+
},
|
| 5842 |
+
{
|
| 5843 |
+
"epoch": 0.9251247920133111,
|
| 5844 |
+
"grad_norm": 0.3078785836696625,
|
| 5845 |
+
"learning_rate": 0.0001169419596950623,
|
| 5846 |
+
"loss": 0.6303,
|
| 5847 |
+
"step": 1668
|
| 5848 |
+
},
|
| 5849 |
+
{
|
| 5850 |
+
"epoch": 0.9262340543538546,
|
| 5851 |
+
"grad_norm": 0.2661837637424469,
|
| 5852 |
+
"learning_rate": 0.00011676485621683713,
|
| 5853 |
+
"loss": 0.4059,
|
| 5854 |
+
"step": 1670
|
| 5855 |
+
},
|
| 5856 |
+
{
|
| 5857 |
+
"epoch": 0.9273433166943982,
|
| 5858 |
+
"grad_norm": 0.2619323134422302,
|
| 5859 |
+
"learning_rate": 0.00011658769861731584,
|
| 5860 |
+
"loss": 0.3383,
|
| 5861 |
+
"step": 1672
|
| 5862 |
+
},
|
| 5863 |
+
{
|
| 5864 |
+
"epoch": 0.9284525790349417,
|
| 5865 |
+
"grad_norm": 0.3200634717941284,
|
| 5866 |
+
"learning_rate": 0.00011641048746840912,
|
| 5867 |
+
"loss": 0.42,
|
| 5868 |
+
"step": 1674
|
| 5869 |
+
},
|
| 5870 |
+
{
|
| 5871 |
+
"epoch": 0.9295618413754853,
|
| 5872 |
+
"grad_norm": 0.29915183782577515,
|
| 5873 |
+
"learning_rate": 0.00011623322334220038,
|
| 5874 |
+
"loss": 0.4156,
|
| 5875 |
+
"step": 1676
|
| 5876 |
+
},
|
| 5877 |
+
{
|
| 5878 |
+
"epoch": 0.9306711037160288,
|
| 5879 |
+
"grad_norm": 0.36972782015800476,
|
| 5880 |
+
"learning_rate": 0.0001160559068109441,
|
| 5881 |
+
"loss": 0.3698,
|
| 5882 |
+
"step": 1678
|
| 5883 |
+
},
|
| 5884 |
+
{
|
| 5885 |
+
"epoch": 0.9317803660565723,
|
| 5886 |
+
"grad_norm": 0.24870358407497406,
|
| 5887 |
+
"learning_rate": 0.00011587853844706397,
|
| 5888 |
+
"loss": 0.4126,
|
| 5889 |
+
"step": 1680
|
| 5890 |
+
},
|
| 5891 |
+
{
|
| 5892 |
+
"epoch": 0.9328896283971159,
|
| 5893 |
+
"grad_norm": 0.2665979862213135,
|
| 5894 |
+
"learning_rate": 0.000115701118823151,
|
| 5895 |
+
"loss": 0.4097,
|
| 5896 |
+
"step": 1682
|
| 5897 |
+
},
|
| 5898 |
+
{
|
| 5899 |
+
"epoch": 0.9339988907376594,
|
| 5900 |
+
"grad_norm": 0.34804749488830566,
|
| 5901 |
+
"learning_rate": 0.00011552364851196167,
|
| 5902 |
+
"loss": 0.3956,
|
| 5903 |
+
"step": 1684
|
| 5904 |
+
},
|
| 5905 |
+
{
|
| 5906 |
+
"epoch": 0.9351081530782029,
|
| 5907 |
+
"grad_norm": 0.2750212550163269,
|
| 5908 |
+
"learning_rate": 0.00011534612808641603,
|
| 5909 |
+
"loss": 0.3434,
|
| 5910 |
+
"step": 1686
|
| 5911 |
+
},
|
| 5912 |
+
{
|
| 5913 |
+
"epoch": 0.9362174154187465,
|
| 5914 |
+
"grad_norm": 0.249044269323349,
|
| 5915 |
+
"learning_rate": 0.00011516855811959604,
|
| 5916 |
+
"loss": 0.4786,
|
| 5917 |
+
"step": 1688
|
| 5918 |
+
},
|
| 5919 |
+
{
|
| 5920 |
+
"epoch": 0.93732667775929,
|
| 5921 |
+
"grad_norm": 0.29392093420028687,
|
| 5922 |
+
"learning_rate": 0.00011499093918474348,
|
| 5923 |
+
"loss": 0.3028,
|
| 5924 |
+
"step": 1690
|
| 5925 |
+
},
|
| 5926 |
+
{
|
| 5927 |
+
"epoch": 0.9384359400998337,
|
| 5928 |
+
"grad_norm": 0.2747836112976074,
|
| 5929 |
+
"learning_rate": 0.00011481327185525828,
|
| 5930 |
+
"loss": 0.4296,
|
| 5931 |
+
"step": 1692
|
| 5932 |
+
},
|
| 5933 |
+
{
|
| 5934 |
+
"epoch": 0.9395452024403772,
|
| 5935 |
+
"grad_norm": 0.3494579493999481,
|
| 5936 |
+
"learning_rate": 0.00011463555670469657,
|
| 5937 |
+
"loss": 0.3412,
|
| 5938 |
+
"step": 1694
|
| 5939 |
+
},
|
| 5940 |
+
{
|
| 5941 |
+
"epoch": 0.9406544647809207,
|
| 5942 |
+
"grad_norm": 0.28468579053878784,
|
| 5943 |
+
"learning_rate": 0.00011445779430676884,
|
| 5944 |
+
"loss": 0.5185,
|
| 5945 |
+
"step": 1696
|
| 5946 |
+
},
|
| 5947 |
+
{
|
| 5948 |
+
"epoch": 0.9417637271214643,
|
| 5949 |
+
"grad_norm": 0.27110087871551514,
|
| 5950 |
+
"learning_rate": 0.0001142799852353382,
|
| 5951 |
+
"loss": 0.3075,
|
| 5952 |
+
"step": 1698
|
| 5953 |
+
},
|
| 5954 |
+
{
|
| 5955 |
+
"epoch": 0.9428729894620078,
|
| 5956 |
+
"grad_norm": 0.38002222776412964,
|
| 5957 |
+
"learning_rate": 0.00011410213006441827,
|
| 5958 |
+
"loss": 0.6445,
|
| 5959 |
+
"step": 1700
|
| 5960 |
+
},
|
| 5961 |
+
{
|
| 5962 |
+
"epoch": 0.9439822518025514,
|
| 5963 |
+
"grad_norm": 0.27994948625564575,
|
| 5964 |
+
"learning_rate": 0.00011392422936817166,
|
| 5965 |
+
"loss": 0.3741,
|
| 5966 |
+
"step": 1702
|
| 5967 |
+
},
|
| 5968 |
+
{
|
| 5969 |
+
"epoch": 0.9450915141430949,
|
| 5970 |
+
"grad_norm": 0.26837414503097534,
|
| 5971 |
+
"learning_rate": 0.00011374628372090783,
|
| 5972 |
+
"loss": 0.3902,
|
| 5973 |
+
"step": 1704
|
| 5974 |
+
},
|
| 5975 |
+
{
|
| 5976 |
+
"epoch": 0.9462007764836384,
|
| 5977 |
+
"grad_norm": 0.2525213062763214,
|
| 5978 |
+
"learning_rate": 0.00011356829369708146,
|
| 5979 |
+
"loss": 0.397,
|
| 5980 |
+
"step": 1706
|
| 5981 |
+
},
|
| 5982 |
+
{
|
| 5983 |
+
"epoch": 0.947310038824182,
|
| 5984 |
+
"grad_norm": 0.24402830004692078,
|
| 5985 |
+
"learning_rate": 0.00011339025987129032,
|
| 5986 |
+
"loss": 0.349,
|
| 5987 |
+
"step": 1708
|
| 5988 |
+
},
|
| 5989 |
+
{
|
| 5990 |
+
"epoch": 0.9484193011647255,
|
| 5991 |
+
"grad_norm": 0.2694087624549866,
|
| 5992 |
+
"learning_rate": 0.0001132121828182738,
|
| 5993 |
+
"loss": 0.4212,
|
| 5994 |
+
"step": 1710
|
| 5995 |
+
},
|
| 5996 |
+
{
|
| 5997 |
+
"epoch": 0.949528563505269,
|
| 5998 |
+
"grad_norm": 0.24677637219429016,
|
| 5999 |
+
"learning_rate": 0.00011303406311291065,
|
| 6000 |
+
"loss": 0.4076,
|
| 6001 |
+
"step": 1712
|
| 6002 |
+
},
|
| 6003 |
+
{
|
| 6004 |
+
"epoch": 0.9506378258458126,
|
| 6005 |
+
"grad_norm": 0.23484551906585693,
|
| 6006 |
+
"learning_rate": 0.00011285590133021741,
|
| 6007 |
+
"loss": 0.3533,
|
| 6008 |
+
"step": 1714
|
| 6009 |
+
},
|
| 6010 |
+
{
|
| 6011 |
+
"epoch": 0.9517470881863561,
|
| 6012 |
+
"grad_norm": 0.2949685752391815,
|
| 6013 |
+
"learning_rate": 0.00011267769804534647,
|
| 6014 |
+
"loss": 0.4117,
|
| 6015 |
+
"step": 1716
|
| 6016 |
+
},
|
| 6017 |
+
{
|
| 6018 |
+
"epoch": 0.9528563505268997,
|
| 6019 |
+
"grad_norm": 0.9004955887794495,
|
| 6020 |
+
"learning_rate": 0.00011249945383358414,
|
| 6021 |
+
"loss": 0.4805,
|
| 6022 |
+
"step": 1718
|
| 6023 |
+
},
|
| 6024 |
+
{
|
| 6025 |
+
"epoch": 0.9539656128674432,
|
| 6026 |
+
"grad_norm": 0.33412787318229675,
|
| 6027 |
+
"learning_rate": 0.00011232116927034893,
|
| 6028 |
+
"loss": 0.5482,
|
| 6029 |
+
"step": 1720
|
| 6030 |
+
},
|
| 6031 |
+
{
|
| 6032 |
+
"epoch": 0.9550748752079867,
|
| 6033 |
+
"grad_norm": 0.24728718400001526,
|
| 6034 |
+
"learning_rate": 0.00011214284493118948,
|
| 6035 |
+
"loss": 0.329,
|
| 6036 |
+
"step": 1722
|
| 6037 |
+
},
|
| 6038 |
+
{
|
| 6039 |
+
"epoch": 0.9561841375485303,
|
| 6040 |
+
"grad_norm": 0.3215670883655548,
|
| 6041 |
+
"learning_rate": 0.00011196448139178298,
|
| 6042 |
+
"loss": 0.4933,
|
| 6043 |
+
"step": 1724
|
| 6044 |
+
},
|
| 6045 |
+
{
|
| 6046 |
+
"epoch": 0.9572933998890738,
|
| 6047 |
+
"grad_norm": 0.21227394044399261,
|
| 6048 |
+
"learning_rate": 0.00011178607922793307,
|
| 6049 |
+
"loss": 0.3171,
|
| 6050 |
+
"step": 1726
|
| 6051 |
+
},
|
| 6052 |
+
{
|
| 6053 |
+
"epoch": 0.9584026622296173,
|
| 6054 |
+
"grad_norm": 0.22638051211833954,
|
| 6055 |
+
"learning_rate": 0.0001116076390155682,
|
| 6056 |
+
"loss": 0.4248,
|
| 6057 |
+
"step": 1728
|
| 6058 |
}
|
| 6059 |
],
|
| 6060 |
"logging_steps": 2,
|
|
|
|
| 6074 |
"attributes": {}
|
| 6075 |
}
|
| 6076 |
},
|
| 6077 |
+
"total_flos": 4209116939550720.0,
|
| 6078 |
"train_batch_size": 8,
|
| 6079 |
"trial_name": null,
|
| 6080 |
"trial_params": null
|